HAL Id: hal-03065412
https://hal.archives-ouvertes.fr/hal-03065412
Submitted on 14 Dec 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Temperature Rise under Two-Photon Optogenetic Brain
Stimulation
Alexis Picot, Soledad Dominguez, Chang Liu, I-Wen Chen, Dimitrii Tanese,
Emiliano Ronzitti, Pascal Berto, Eirini Papagiakoumou, Dan Oron, Gilles
Tessier, et al.
To cite this version:
Alexis Picot, Soledad Dominguez, Chang Liu, I-Wen Chen, Dimitrii Tanese, et al.. Temperature
Rise under Two-Photon Optogenetic Brain Stimulation. Cell Reports, Elsevier Inc, 2018, 24 (5),
pp.1243-1253.e5. �10.1016/j.celrep.2018.06.119�. �hal-03065412�
Article
Temperature Rise under Two-Photon Optogenetic
Brain Stimulation
Graphical Abstract
Highlights
d
In vivo multi-target 2P optogenetics might induce local
sample heating
d
A model is developed to predict and minimize sample heating
d
The model can predict temperature rise under scanning and
holographic illumination
d
The model is validated using erbium-ytterbium co-doped
glass particles
Authors
Alexis Picot, Soledad Dominguez,
Chang Liu, ..., Gilles Tessier,
Beno^ıt C. Forget, Valentina Emiliani
Correspondence
valentina.emiliani@parisdescartes.fr
In Brief
Picot et al. model light and heat diffusion
under the typical illumination conditions
used for single- and multi-target
two-photon optogenetics and compare the
heat distribution under parallel and spiral
scanning illumination. They
experimentally validate the model using
the temperature-dependent fluorescence
of erbium-ytterbium co-doped glass
particles.
Picot et al., 2018, Cell Reports24, 1243–1253 July 31, 2018ª 2018 The Authors.
Cell Reports
Article
Temperature Rise under Two-Photon
Optogenetic Brain Stimulation
Alexis Picot,1Soledad Dominguez,1Chang Liu,2,3I-Wen Chen,1Dimitrii Tanese,1Emiliano Ronzitti,1,3Pascal Berto,2
Eirini Papagiakoumou,1,4Dan Oron,5Gilles Tessier,2,3Beno^ıt C. Forget,1and Valentina Emiliani1,6,*
1Wavefront-Engineering Microscopy Group, Neurophotonics Laboratory, UMR 8250 CNRS, University Paris Descartes, 45 rue des Saints-Pe`res, 75006 Paris, France
2Holographic Microscopy Group, Neurophotonics Laboratory, UMR 8250 CNRS, University Paris Descartes, 45 rue des Saints-Pe`res, 75006 Paris, France
3Sorbonne Universite´, CNRS, INSERM, Institut de la Vision, 17 Rue Moreau, 75011 Paris, France 4Institut National de la Sante´ et de la Recherche Me´dicale (Inserm), Paris, France
5Department of Physics of Complex Systems, Weizmann Institute of Science, Rehovot 76100, Israel 6Lead Contact
*Correspondence:valentina.emiliani@parisdescartes.fr https://doi.org/10.1016/j.celrep.2018.06.119
SUMMARY
In recent decades, optogenetics has been
transform-ing neuroscience research, enabltransform-ing neuroscientists
to drive and read neural circuits. The recent
develop-ment in illumination approaches combined with
two-photon (2P) excitation, either sequential or parallel,
has opened the route for brain circuit manipulation
with single-cell resolution and millisecond temporal
precision. Yet, the high excitation power required
for multi-target photostimulation, especially under
2P illumination, raises questions about the induced
local heating inside samples. Here, we present and
experimentally validate a theoretical model that
makes it possible to simulate 3D light propagation
and heat diffusion in optically scattering samples
at high spatial and temporal resolution under the
illu-mination configurations most commonly used to
perform 2P optogenetics: single- and multi-spot
ho-lographic illumination and spiral laser scanning. By
investigating the effects of photostimulation
repeti-tion rate, spot spacing, and illuminarepeti-tion dependence
of heat diffusion, we found conditions that make it
possible to design a multi-target 2P optogenetics
experiment with minimal sample heating.
INTRODUCTION
Over the past 15 years, optogenetics has become a unique and powerful tool for the investigation of brain circuits, with the capa-bility of controlling neuronal firing and inhibition with millisecond precision and cell specificity (Emiliani et al., 2015).
Wide-field single-photon (1P) illumination is the most commonly used method to activate optogenetic actuators ( Boy-den et al., 2005; Nagel et al., 2005). Combined with strategies that restrict opsin expression in specific neuronal sub-popula-tions (Beltramo et al., 2013; Cardin et al., 2009; Kuhlman and
Huang, 2008) and/or optical fibers (Aravanis et al., 2007; Penzo et al., 2015; Wu et al., 2014) to reach deep brain regions, 1P wide-field illumination has been widely applied in neuroscience research, for example, inducing synaptic plasticity (Zhang and Oertner, 2007), mapping brain circuitry (Adesnik et al., 2012), and modulating behaviors (Adamantidis et al., 2007; Huber et al., 2008; Kitamura et al., 2017). Recently, additional strategies for 1P patterned illumination have made it possible to further in-crease photostimulation precision by restricting the illumination on specific brain layers (Pisanello et al., 2014) or cellular and sub-cellular structures (Guo et al., 2009; Petreanu et al., 2009; Szabo et al., 2014; Wyart et al., 2009). Yet the main constraint of 1P-based illumination approaches remains the limited spatial resolution and penetration depth: wide-field illumination makes it impossible to target an individual cell within a dense neuronal ensemble, while 1P patterned approaches have reached only shallow depths (Szabo et al., 2014).
This has spurred the development of more sophisticated light-delivering methods using two-photon (2P) excitation. These can be divided into two main categories: scanning approaches (Packer et al., 2012, 2015; Prakash et al., 2012; Rickgauer and Tank, 2009; Yang et al., 2018), in which cell photocurrent builds up thanks to the sequential photostimulation of channels or pumps expressed at the target membrane, and parallel ap-proaches (low numerical aperture Gaussian beam, generalized phase contrast, computer-generated holography [CGH], tempo-ral focusing), in which this is achieved by simultaneous illumina-tion of the entire target (Be`gue et al., 2013; Chaigneau et al., 2016; dal Maschio et al., 2017; Papagiakoumou et al., 2010; Rickgauer et al., 2014; Ronzitti et al., 2017). The combination of these approaches with ad hoc engineered control tools has enabled in-depth optical control of neuronal firing with milli-second temporal precision and cellular resolution (Ronzitti et al., 2017). Very recently, 2P illumination methods have been extended to the generation of three-dimensional (3D) illumination patterns. This has been achieved by spiral scanning multiplexed holographic foci (Packer et al., 2015; Yang et al., 2018) or by simultaneous shining of multiple spatiotemporally focused spots (Accanto et al., 2017; Hernandez et al., 2016; Pegard et al., 2017; Sun et al., 2018). These approaches combined with the use of
high-energy fiber lasers (Chaigneau et al., 2016; Ronzitti et al., 2017; Yang et al., 2018) and eventually highly sensitive opsins make it theoretically possible to simultaneously target hundreds of cells within cubic millimeter-size illumination volumes (i.e., comparable with what is achievable with single-fiber visible light illumination), preserving at the same time the single-cell resolu-tion of 2P illuminaresolu-tion. Yet setting the real limitaresolu-tions for multi-target 2P illumination requires considering possible sources of photodamage. These include both linear effects, such as thermal damage related to the linear absorption of light, and nonlinear, multiphoton absorption processes, inducing photochemical, ablation damage or optical breakdown (Hopt and Neher, 2001; Koester et al., 1999; Vogel et al., 2005) arising at peak fluences of about 0.1 J/cm2 for Chinese hamster ovarian cells (Ko¨nig
et al., 1999), 0.5–2 J/cm2in water (Linz et al., 2016; Noack and Vogel, 1999; Vogel et al., 2005), and 1.5–2.2 J/cm2for porcine corneal stroma (Olivie´ et al., 2008).
Because of the short dwell time and small illumination volume used in conventional multi-photon imaging, heating through linear absorption can be considered a negligible source of pho-todamage (De´barre et al., 2014; Kobat et al., 2009; Koester et al., 1999; Linz et al., 2016). Nevertheless, for repetitive scanning of large areas, this can become of increasing importance (Hopt and Neher, 2001; Podgorski and Ranganathan, 2016). Unlike 2P imaging, parallel optogenetic neuronal activation uses long (milliseconds to seconds) exposure time and/or large illumina-tion area (or amount of targets), so thermal phenomena require a careful evaluation (Boulnois, 1986). Many neural functions can be altered when there is a change of temperature (Aronov and Fee, 2012; Christie et al., 2013; Elwassif et al., 2006; Kalm-bach and Waters, 2012; Wang et al., 2014). Even small temper-ature changes can cause modulations of the action potential (AP) shape (Hodgkin and Katz, 1949), firing rate of neurons (Reig et al., 2010; Stujenske et al., 2015), channel conductance ( Shiba-saki et al., 2007; Wells et al., 2007), or fluctuation of the synaptic responses (Andersen and Moser, 1995; Thompson et al., 1985). Cell death after denaturation of proteins can also be expected af-ter a temperature increase of 6–8 K above the physiological resting temperature (Deng et al., 2014; Thomsen, 1991). Notably, thermal damage thresholds also depend on brain area and spe-cific tissue properties (Kiyatkin, 2007; Sharma and Hoopes, 2003).
Temperature rise under the typical illumination conditions for 1P optogenetics (i.e., wide-field illumination through optical fibers and long, 0.5–60 s exposure time) has been investigated both theoretically, using Monte Carlo with finite-difference time-domain schemes (Stujenske et al., 2015) or the finite-element method (Shin et al., 2016), and experimentally using thermocouples (Shin et al., 2016; Stujenske et al., 2015), quan-tum dots (Podgorski and Ranganathan, 2016), or infrared (IR) cameras (Arias-Gil et al., 2016).
Recently, Podgorski and his colleagues have modeled and measured heating under 20–180 s 2P repeated scanning illumi-nation of a volume measuring 1 mm2in surface and 250mm in depth (Podgorski and Ranganathan, 2016).
Despite these first investigations, a careful evaluation of brain heating under the different illumination conditions (parallel and scanning) used for 2P optogenetics is still missing.
In this paper, we present a theoretical model to describe light propagation and heat diffusion with millisecond precision and micrometer resolution under typical 2P excitation conditions that enable in vivo optogenetic control of neuronal firing. The model combines a random phase mask approach to account for 3D light scattering within tissue and Fourier’s heat diffusion equation solved through Green’s function formalism to evaluate the corresponding spatial and temporal heat diffusion during and after 2P photostimulation. We validated the model by comparing simulated and measured laser-induced temperature changes in a water/agar gel using the temperature dependent fluorescence emission of erbium-ytterbium (Er/Yb) co-doped glass particles. We then use the model to predict the temporal and spatial heat distribution in scattering media under different illumination conditions, including single- and multi-spot holographic illumi-nation and spiral scanning. We analyze the 3D spatial and tem-poral evolution of temperature rise as a function of the stimula-tion frequency, laser repetistimula-tion rate, and illuminastimula-tion durastimula-tion.
The model is extendable to other illumination configurations, brain structures, and biological preparations with different scattering properties and is a unique and powerful tool to design the optimal illumination conditions for 2P optogenetics brain circuit control. The model has been implemented in a MATLAB (The MathWorks) package for use by other users to predict heat diffusion under their own 2P optogenetics experimental conditions.
RESULTS
Modeling Heat Diffusion
Heating during photostimulation (or any experimental procedure involving shining light on or in an object) results from the thermal-ization of the energy from the light source absorbed by the tissue.
To model heat diffusion, we considered brain tissue as a uni-form and isotropic medium initially at temperature T0,
character-ized by a thermal diffusion constant (or diffusivity) D, a specific heat C,and density r (seeTable S1). The spatiotemporal distribu-tion of the temperature rise Tð r!;tÞ, where r! is the spatial coor-dinate in three dimensions and t is time, is obtained from Fouri-er’s heat diffusion equation (Fourier, 1822):
vTð r!; tÞ
vt = DV2Tð r!;tÞ+qð r !;tÞ
rC ; (1)
where qð r!;tÞ=fð r!;tÞma is the heat source term corresponding to the deposited energy flux per unit of volume from the absorp-tion of laser power,fð r!; tÞ is the irradiance, and ma is the ab-sorption coefficient of the medium. In the 950–1,030 wavelength range, water is the main source of absorption, while the contribu-tion from hemoglobin is almost negligible. Therefore for all simu-lations (in vitro and in vivo), we used for ma the ex vivo value given inYaroslavsky et al. (2002), which in this wavelength regime is very close to the in vivo situation (Johansson, 2010).
In living tissue, it is common to use the so-called Pennes bio heat equation (Pennes, 1948), which would add to the previ-ous equation a source related to metabolic process, qm, and a
sink (cooling) term related to blood perfusion, qp, within the tissue: vTð r!; tÞ vt = DV2Tð r!;tÞ+qð r !;tÞ rC + qm qp rC : (2)
The sink term is a function of the thermal properties of blood (rb, Cb), blood flow wb, and temperature Tband is expressed as qp= rbCbwbððT0+ Tð r!;tÞÞ TbÞ. In physiological conditions
qm and qp should be equal, maintaining tissue temperature constant.
If we take the characteristic values forrb, Cb, and wb(Elwassif
et al., 2006; Stujenske et al., 2015) and we consider a tissue temperature rise of 1 K, we obtain for the sum of the source and sink term,qm qp
rC , a value of roughly9.2 3 103K/s1. By taking for qð r!;tÞ under our experimental conditions a value of6 3 106mW/mm3(withfð r!; tÞ 0.1 mW/mm2and
ma = 63 105mm1), we obtain forqð rrC!;tÞa value of roughly 1.73 103K/s1, withr and C taken fromBlumm and Lindemann (2003)andYizhar et al. (2011).
In agreement with previous findings (Elwassif et al., 2006; Stujenske et al., 2015), we can consider that under the experi-mental conditions considered in this paper, the main cooling mechanism is through diffusion. We therefore neglected the
qm qp
rC term and used the Fourier’s heat diffusion,Equation 1. To solveEquation 1, we used Green’s function formalism ( Car-slaw and Jaeger, 1947). Green’s function Gð r!;tÞ is the solution to an instantaneous point source of heat, and Tð r!;tÞ is given by the convolution (over space and time) of this Green’s function and the source term qð r!; tÞ. For an infinite media (seeSTAR Methods), Green’s function is readily obtained in analytical form for one-dimensional (1D) diffusion,
1 ffiffiffiffiffiffiffiffiffiffiffi 4pDt p exp x2 4Dt ; (3) or 3D diffusion, 1 ð4pDtÞ3=2exp r2 4Dt ; (4) where r2= x2+ y2+ z2.
As Green’s function for the diffusion equation is a Gaussian distribution, it is common, as seen for example inBird et al. (1976), to define a diffusion length (here a thermal diffusion length) as the SD of this Gaussian distribution: lth=
ffiffiffiffiffiffiffiffiffiffiffi 2nDt p
, where n = 1, 2, 3 is the dimensionality of the medium in which the diffusion process occurs.
Convolution of Green’s function and the source term must then be carried out over space and time. In order to facilitate this convolution, we expressed the source term as a separable function of space and time, meaning as the product of the func-tion describing independently the spatial and temporal distribu-tions: qð r!; tÞ = fð r!; tÞma = Gð r!ÞPðtÞma. Spatial distribution
Gð r!Þ and time dependence PðtÞ of the source term can therefore be treated separately. Further details describing the modeling of the light source propagation, the scattering, the time dependence of source term, and the trajectory of the laser beam can be found in theSTAR Methods.
Experimental Validation of the Model
We tested the accuracy of our model by comparing the theoret-ical prediction with the measured temperature rise induced by a holographic spot focalized on a water/agar gel by embedding a micrometric Er/Yb co-doped glass particle in the gel (Figures S1A and S1B) and recording the luminescence spectral changes on the particle by the laser-induced heating (see STAR Methods).
We started by comparing simulations, with water coefficient of absorption taken fromKedenburg et al. (2012), with experiments in the case of a 500 ms stimulation at 1,030 nm wavelength (laser pulse width 250 fs, repetition rate 10 MHz) focalized to a 15-mm-diameter holographic spot placed 30 mm away from the particle, and we found that our prediction perfectly reproduced both the magnitude and the temporal evolution of the tempera-ture rise (Figure 1A, left). To validate the ability of the model to account for the spatial heat distribution, we compared the pre-dicted and experimental values of the temperature rise reaching 500 ms after optical excitation, while laterally moving the illumination spot with respect to the particle (Figure 1A, right). Finally, we exploit the capability of the model to predict the fast temperature changes and temperature accumulation produced by 2 or 10 Hz stimulation trains (Figure 1B) of 50 ms illumination pulses.
Single-Spot Holographic Photostimulation
Here, we use the model to predict the temperature changes produced under exemplary illumination condition for 2P opto-genetics using parallel (holographic) illumination (Chen et al., 2017).Figure 2A (left) shows the simulated temperature change produced by a single holographic spot after propagation through 150 mm of brain tissue (see STAR Methodsand Fig-ure S2A) using exemplary conditions for in vivo AP generation (Figure 2A, right; seeSTAR Methods) using the opsin CoChR. At the end of the 3 ms illumination time, the model predicts a mean temperature increase of the cell of 0.3 K followed by a rapid temperature decay (reaching 0.05 K after 10 ms). It should be noted that although the temperature increase has a linear dependence with the excitation power, it is not linear with exposure time when diffusion is taken into account. This sublinear dependence can be expressed in terms of the com-plementary error function, and for the typical conditions consid-ered in this paper, increasing the exposure time of a factor of 10 raises the temperature only by a factor of roughly 2 ( Fig-ure S3B). Although at the onset of heat diffusion, the tempera-ture distribution reproduces the speckled intensity distribution typical of CGH, these fluctuations are washed out as soon as the thermal diffusion length ðlth=
ffiffiffiffiffiffiffiffi 6Dt p
Þ equals the speckle size (lthzl = 1; 03 mm; i.e., for tR1 ms) (Figure S3A). We then simulated the effect on thermal response when the same stim-ulation (illumination time = 3 ms) was repeated at a rate of 10 or 40 Hz (Figure 2B, left) to produce AP trains (Figure 2B, right). As
we can observe, under 10 Hz stimulation, the heat dissipation after each photostimulation pulse is fast enough to bring the cell back to the equilibrium temperature before the arrival of the next photostimulation pulse. Increasing the stimulation repetition rate generates a heat accumulation after each photo-stimulation pulse. However, even at a rate of 100 Hz, after five pulses, the accumulated temperature rise does not exceed a few tenths of a degree (Figure 2C); of course this accu-mulation becomes more relevant for longer illumination times (Figure S3B).
Multi-spot Holographic Photostimulation
Here, we consider the temperature rise induced by multiple spots distributed in a volume. In this case, it is important to consider for each plane the different attenuation of light due to scattering. To this end, the light distribution is obtained by using a holographic phase mask that compensates for the depth-dependent light los-ses and produces, at each focal depth, holographic spots of equal excitation density. Using a 3D fluorescence stack as a guide ( Fig-ure 3A) for placing the spots in three dimensions, we generated 100 holographic spots within a 3003 300 3 300 mm3volume (excitation power density at each spot 0.07 mW/mm2) and used our model to predict the corresponding 3D spatiotemporal evolution of the temperature rise (Figure 3B;Video S1). At the end of the 3 ms stimulation, we found a mean temperature rise, averaged over the 100 spots, of1 K (Figure 3C). Because of the high density of spots, the value of the temperature rise at each spot obviously depends on the number and locations of the neighboring spots. For example, at spots generated in the deeper layers and surrounded by several neighboring spots, the
A B 50 ms - 2 Hz 50 ms - 10 Hz Er/Yb doped particle X Holographic spot
Figure 1. Validation of the Thermal Simula-tions through Experimental Luminescence Recordings
(A) Left: simulation (black trace) and experimental measurement (blue trace) of the temperature rise induced by a 15-mm-diameter holographic spot (500 mW average power, 500 ms illumination time, 1,030 nm excitation wavelength, 10 MHz repetition rate,250 fs laser pulse duration) placed at 30 mm from an Er/Yb co-doped particle placed in water/ agar gel (mass fraction of agar 0.005) at a depth of 150mm. Right: simulation and experimental mea-surement of the peak temperature rise as a function of the distance X between the particle and the ho-lographic spot using the same illumination condi-tions as in (A, left). Scale bar, 10mm.
(B) Simulation (black trace) and experimental measurement (blue trace) of the temperature rise induced by a 2 Hz (left) and 10 Hz (right) train of 50 ms illumination pulses, using the same illumi-nation conditions as in (A) with the spot placed at 30mm from the particle.
local temperature rise can reach up to 1.85 K (Figure 3C), while for spots placed at an average distance greater than the thermal diffusion length (lth=
ffiffiffiffiffiffiffiffi 6Dt p
50 mm) from their nearest neighbor, the temperature rise remains around 0.3 K (i.e., comparable with the case of the isolated spot reported inFigure 2A).
Simulation of Heating Effect for Different Illumination Conditions
Here, we show how our model can simulate the spatiotemporal temperature distribution generated under spiral scanning illumi-nation and compare the corresponding heat distribution with the one obtained using holographic illumination.
To quantitatively compare the two approaches, we first deter-mined for each of these two approaches the power conditions that allow in vitro AP generation with comparable latencies. To this end, we used an optical system able to perform sequentially holographic and spiral scanning photostimulation on the same cell (see STAR Methods). We found that for both short (3 ms) and long (40 ms) illumination times, the average power neces-sary to evoke an AP using holographic illumination was roughly twice larger than with the scanning approach (37.5 and 16 mW on average [n = 3] for 3 ms illumination, 5.8 and 2.5 mW on average [n = 4] for 40 ms illumination;Figure 4A). Of note, the larger spot in holographic excitation enables the use of a power density roughly 150 times smaller (0.2 versus 31 mW/mm2 for 3 ms illumination, 0.03 versus 5 mW/mm2for 40 ms illumination).
These results confirm what has been recently achieved in vivo using the C1V1 opsin and a laser source tuned to 1 MHz (average power for spiral scanning about 1.8 lower than the one used for a 12-mm-diameter holographic spot) (Yang et al., 2018).
These power values were used to simulate the spatiotem-poral distribution of the temperature rise using holographic
(Figure 4B) and spiral scanning illuminations (Figure 4C) and two illumination durations (3 and 40 ms). The maximum mean temperature is reached, as expected, at the end of the illumina-tion time. After 3 ms of illuminaillumina-tion (Figure 4B, left;Figure 4C, left;Video S3), it equals1.1 K for CGH and 0.5 K for scan-ning. In the latter case, the accumulation of heat during the scan leads to a localized temperature peak of1.1 K (about 2 times higher than the average value) when the laser reaches the center of the spiral. Also, the model makes it possible to predict temperature oscillations at the edge of the cell during the seven scanning loops. Using longer illumination times (40 ms) makes it possible to decrease the excitation power, although it also increases the AP latency; in this case the mean temperature stays below 0.25 and 0.12 K for CGH
A
B
C
Figure 2. Simulated Temperature Rise Pro-duced by a Holographic Spot Using the Photostimulation Conditions Necessary to EvokeIn Vivo Action Potential
(A) Left: temperature rise, averaged on the spot surface, produced by a 12-mm-diameter holo-graphic spot at a depth of 150mm during 3 ms illumination, 1,030 nm excitation wavelength, 500 kHz repetition rate, 250 fs laser pulse duration, 11.3 mW excitation average power at the objec-tive focal plane corresponding to0.07 mW/ mm2
at a depth of 150mm (and to 0.1 mW/ mm2
exci-tation power density in the absence of scattering). Right: in vivo voltage recordings in cell-attached configuration from a CoChR-expressing cortical neuron upon photostimulation with the parame-ters used for the simulation in (A, left) evoking an action potential.
(B) Left: temperature rise, averaged on the spot surface, produced by a 12-mm-diameter holo-graphic spot during a train of five illuminations of 3 ms at 10 or 40 Hz using the conditions described in (A). Right: experimental voltage re-cordings in cell-attached configuration in vivo from a CoChR-expressing cortical neuron upon photostimulation using the same parameters as in the left panel. Action potentials were induced upon each photostimulation with trains of 3 ms light pulse (red mark).
(C) Temperature rise, averaged on the spot sur-face, produced by a 12-mm-diameter holographic spot during a train of five illuminations pulses of 3 ms at 100 Hz.
and scanning, respectively, with a local maximum of 0.2 K at the center of the cell in the case of spiral scanning (Figure 4B, right;Figure 4C, right).
DISCUSSION
In this study, we provide a method able to measure and predict the temperature changes induced by 2P illumination with micrometer precision and milli-second temporal resolution, under the typical condition used for in vitro and in vivo 2P optogenetics stimulation.
Combining a random phase mask approach with the Fourier’s heat diffusion equation to account for 3D light scattering and heat diffusion within tissue, the model makes it possible to follow the 3D spatiotemporal evolution of temperature rise both under single and multi-target activation.
Previous approaches such as Monte Carlo simulations with finite-difference time-domain (FDTD) methods (Stujenske et al., 2015) or empirical fitting of experimental results (Arias-Gil et al., 2016; Podgorski and Ranganathan, 2016) did not reach micrometer and millisecond resolution. Indeed, to simulate the propagation of tightly focused beams would require introducing diffraction in the Monte Carlo code, which is intrinsically difficult
(Brandes et al., 2014). Furthermore, for holographic beams, a Monte Carlo approach would need to simulate the propagation of the electromagnetic field, which is much more complex than the traditional Monte Carlo schemes on the basis of the propaga-tion of the intensity. On the other hand, FDTD methods require extra care in setting the spatial and time discretization steps in order to assure stability and avoid spurious oscillations in the nu-merical solution. For a Crank-Nicolson scheme (Crank and Nic-olson, 1947), this required satisfying the conditionDDt
Dx2< 1 2with D
the thermal diffusivity, which bounds the spatial and the tempo-ral resolution. For example, following the time interval of 3 ms with a micrometer resolution ofDx = 0:5 mm would imply setting Dtz1 ms and therefore using 3,000 iterations. Other numerical schemes can be used to solve the heat diffusion equation, but in all cases, they are iterative processes and therefore are by themselves time consuming, particularly if high accuracy is needed. Empirical fitting of temperature measurements is intrin-sically limited to the experimental spatial and temporal resolu-tion, which so far were significantly above the micrometer and millisecond scale.
On the other hand, using the analytical expression of the Green’s function gives us the possibility to simulate the spatial distribution of temperature with micrometer spatial precision and submillisecond temporal resolution and discuss the results from the simulation on the basis of a physical parameter that is the thermal diffusion length, lth.
Figure 3. Simulated Temperature Rise Pro-duced by Multiple Holographic Spots at the Photostimulation Conditions Necessary to EvokeIn Vivo Action Potential
(A) Three-dimensional view of an in vivo two-photon fluorescence stack of the layer 2/3 of mouse visual cortex labeled with GFP with in red locations of the 100 holographic spots.
(B) Three-dimensional spatial distribution of irradi-ance produced by 100 holographic spots (12mm diameter, 0.1 mW/mm2
, 1,030 nm) placed in a 3003 3003 300 mm3
volume, with scattering compen-sated for in vivo conditions.
(C) Three-dimensional spatial distribution of the temperature rise produced by 100 holographic spots (12 mm diameter, 0.1 mW/mm2
, 1,030 nm) placed in a 3003 300 3 300 mm3
volume, after 2 ms of illumination.
(D) Examples of traces of the temporal evolution of the temperature rise, averaged on the spot surface, produced by 3 ms of illumination with the spots distribution described in (B) and the illumination conditions described inFigure 2for an isolated spot (black curve) and a spot chosen in the area with high density of spots (blue curve). The purple curve represents the meant temperature rise averaged over the 100 spots.
So far, experimental methods to evaluate light induced heating during optogenetics experiments have used thermocouples, with millimeter-long tips and diameters ranging between 220 and 500 mm, which translates into a millimeter-range sensitive region (Podgorski and Ranganathan, 2016; Shin et al., 2016; Stujenske et al., 2015), or a thermal camera allowing a spatial resolution of 50 mm (Arias-Gil et al., 2016). These probes were well adapted to measure temperature rises averaged on large area but lack the necessary precision to predict the spatial heat distribution induced by 2P patterned light at micro-meter scale. The use of quantum dot thermometry could reduce the spatial resolution down to the micrometer scale but required long (typically 1 s) integration time (Podgorski and Ranganathan, 2016).
Here, we quantify (at high spatiotemporal resolution) the tem-perature response induced by 2P excitation by using rare-earth doped glass particles. Such particles have the property of emit-ting a strong temperature-dependent luminescent signal (Aigouy et al., 2005; Saı¨di et al., 2009). By probing a single10 mm Er/Yb co-doped particle, we manage to significantly increase the photon budget to reach a thermal sensitivity below 0.2 K, a tem-poral resolution of 4 ms, while reducing the size of the probed re-gion down to a spatial scale comparable with the neuron cell body. To efficiently sample the temporal evolution of heating, we used here illumination pulses > 50 ms. More efficient detec-tion schemes will increase sensitivity, enabling higher sampling rates and thus the use of shorter illumination pulses. However, the Er/Yb co-doped particle lifetime will limit the resolution to 0.5 ms (Wang et al., 2016).
After experimental verification of the validity of the model, we have used this to evaluate the spatiotemporal heat distribution un-der the most commonly used configurations for 2P optogenetics control of neuronal firing: holographic and scanning illumination.
Holographic parallel illumination combined with amplified low-repetition rate lasers and sensitive opsins makes it possible to evoke AP and AP trains in vivo using low excitation density (<0.1 mW/mm2) and short illumination time (<3 ms) (Chen et al., 2017). Using our model we have shown that these illumination conditions correspond to less than 0.35 K of mean temperature increase for single-cell activation (Figure 2).
In 2P CGH, the intensity distribution of the excitation patterns present spatial intensity fluctuations that could reach 50% around the mean value (Papagiakoumou et al., 2008), which could generate localized hotspots. However, we could show that this speckled distribution is preserved in the heating profile only for the first few microseconds, being quickly smoothed out by diffusion. Within this short time, the temperature rise even at
Figure 4. Simulated Temperature Rise and Experimental Electrophysiological Record-ings with CGH and Scanning Illumination Techniques
(A) Left: excitation average power needed to generate in CoChR-expressing neuronal brain sli-ces a single AP with a latency between 2 and 10 ms using a 15-mm-diameter holographic spot illumi-nation for 3 ms, or spiral scanning (seven revolu-tions moving from the edge to the center, 15mm diameter) for the same duration, 3 ms (laser exci-tation 1,030 nm, 2 MHz repetition rate,300 fs laser pulse duration). Right: excitation power needed to generate in CoChR-expressing cell in acute brain slices a single AP with a latency be-tween 20 and 45 ms using a holographic illumina-tion for 40 ms, or spiral scanning (seven revoluillumina-tions moving from the edge to the center, 15mm diam-eter) for the same duration, 40 ms (laser excitation 1,030 nm, 2 MHz repetition rate,300 fs laser pulse duration).
(B) Left: temperature rise averaged on the spot surface, produced using the holographic illumina-tion condiillumina-tion of the experiments of (A, left) and 37.5 mW excitation average power. Right: tem-perature rise averaged on the spot surface, pro-duced using the holographic illumination condition of the experiments of (B, right) and 5.8 mW average power.
(C) Left: Temperature rise, using the spiral illumi-nation condition of the experiments of (A, left) and 16 mW excitation average power. Right: temper-ature rise, produced using the spiral illumination conditions of the experiments of (B, right) and 2.5 mW average power. In blue, temperature rise at the beginning, on the edge. In black, the temper-ature rise at the center of the spiral. In red, the temperature rise averaged over a disc of 15mm.
the hottest spots will not exceed a few millikelvin, thus ruling out the risk that holographic speckles can induce local hotspots.
The generation of trains of pulses had no effect on the total temperature rise, for a stimulation frequency of 10 Hz and 3 ms illumination time (Figure 2B). For higher stimulation frequency, when the delay between pulses becomes comparable with the heating decay time, repetitive pulse stimulation will induce heat-ing accumulation, but even for the case of 100 Hz stimulation, this will be less than 0.1 K (Figure 2C).
For prolonged (seconds to minutes) scanning illuminations of large area (square millimeters), the highest temperature change is deeper in the brain than at the focal plane (Podgorski and Ran-ganathan, 2016) (see STAR Methods), while for the spatially localized and short (milliseconds) illumination conditions used in 2P optogenetics, the highest temperature changes occurred within the spots (Figure 3B), suggesting that these are the posi-tions where it is important to evaluate the maximum temperature rise. We have shown that the simultaneous illumination with 100 spots, placed in a 3003 300 3 300 mm3volume, makes it possible to keep the temperature rise at the targets comparable
with the case of an isolated cell, provided that the spatial dis-tance among the targets is kept larger than the 3D diffusion length (Figure S4;Video S2).
We have shown that our model can be applied to predict tem-perature rise both under parallel and spiral scanning optogenetic illuminations. In scanning approaches, the excitation light is focused on a small spot, and photocurrent builds up thanks to the sequential opening of channels. This enables using lower illu-mination power than the one used in holographic photostimula-tion, for which, on the contrary, current integration is achieved by simultaneous activation of all channels. The smaller spot size used in spiral scanning also enables more efficient heat dissipa-tion, so that overall spiral scanning leads to lower temperature rise. On the other hand, the concentration of light on a small spot leads to power densities 150 times higher than the one used in CGH and closer to nonlinear photodamage thresholds so that, especially for short illumination times, care needs to be taken in limiting the number of successive scans (Hopt and Neher, 2001).
These results also indicate that the optimal laser repetition rate for 2P optogenetics depends on the adopted illumina-tion methods: the extremely low excitaillumina-tion power density adopted for parallel illumination approaches (Figure 4B: peak fluencez 10.7 and 1.7 mJ/cm2, respectively, for 3 and 40 ms illu-mination conditions;Figure 2: peak fluencez 20 mJ/cm2) makes it possible to neglect nonlinear damage effect and privileges using low (500 kHz to 2 MHz) repetition rate lasers to minimize heating through linear absorption. Scanning approaches require higher excitation power density (Figure 4C: peak fluence z 1.6 and 0.25 J/cm2, respectively, for 3 and 40 ms illumination conditions)
but enable more efficient heat dissipation; therefore, for short illumination times, a higher repetition rate laser (Ji et al., 2008) should be preferred in order to minimize peak-power-sensitive damages.
Here, we used for both approaches the opsin CoChR, which has intermediate rise (6 ms) and decay (35 ms) times ( She-mesh et al., 2017). For parallel approaches, similar excitation po-wer density and illumination times can be reached independently on the opsin kinetics (Chaigneau et al., 2016; Chen et al., 2017; Ronzitti et al., 2017); therefore we can expect similar values of temperature rises. On the contrary, illumination conditions for spiral scanning are more sensitive to the opsin kinetics: combi-nation with a slower opsin such as C1V1 makes it possible to lower the power (Yang et al., 2018) and therefore reduce the tem-perature rise even further, while combined with faster opsins as Chronos would require the use of higher powers to compensate for the fast channel off-time.
Our model is extendable to other brain regions or biological preparations differing in scattering properties, thus offering a unique and flexible tool for the design of complex optogenetics experiments with minimal sample heating. It will surely prove useful also to simulate the temperature distribution under different excitation configurations (including single- and three-photon excitation) and imaging geometries (e.g., light sheet mi-croscopy, stimulated emission depletion mimi-croscopy, or Bessel beam illumination) or to optimize light distribution and illumina-tion condiillumina-tions for thermogenetic experiments (Bernstein et al., 2012; Ermakova et al., 2017; Hamada et al., 2008).
STAR+METHODS
Detailed methods are provided in the online version of this paper and include the following:
d KEY RESOURCES TABLE
d CONTACT FOR REAGENT AND RESOURCE SHARING
d EXPERIMENTAL MODEL AND SUBJECT DETAILS
B Mice for in vivo experiments
B Mice for in-vitro experiments
d METHOD DETAILS
B Virus injection and surgical procedures – In-vivo
B 2P-guided electrophysiology – in-vivo
B Holographic photostimulation – in-vivo
B Brain slices
B Whole-cell recordings – in-vitro
B 2P holographic and spiral scanning photostimulation – in-vitro
B Modeling – Light source propagation and scattering
B Modeling – Time dependence of source term
B Modeling – Convolution with separable source term
B Modeling – Moving laser spot
B Modeling – The infinite media hypothesis
B Thermal measurement
B Thermal measurement – Calibration of the probe
d QUANTIFICATION AND STATISTICAL ANALYSIS
d DATA AND SOFTWARE AVAILABILITY SUPPLEMENTAL INFORMATION
Supplemental Information includes four figures, one table, three videos, and one data file and can be found with this article online athttps://doi.org/10. 1016/j.celrep.2018.06.119.
ACKNOWLEDGMENTS
We acknowledge the SCM (Service Commun de Microscopie – Faculte´ des Sciences Fondamentales et Biome´dicales – Paris) for providing the software Imaris (version 8.4; Bitplane;www.bitplane.com) and 3i for technical support in the implementation of spiral scanning. We thank Agence Nationale de la Re-cherche (grants ANR-14-CE13-0016 [Holohub] and ANR-15-CE19-0001-01 [3DHoloPAc]), the Human Frontiers Science Program (grant RGP0015/2016), Re´gion Ile de France (Projet GeneTherm – C’Nano – DIM Nano-K 2016), Fon-dation Bettencourt Schueller (Prix Coups d’e´lan pour la recherche´ franc¸aise), and the Getty Lab for financial support. I.-W.C. received funding from the Eu-ropean Union’s Horizon 2020 research and innovation program under Marie Sk1odowska-Curie grant agreement 747598. We thank Patrick Gredin and Mi-chel Mortier, from Chimie Paris, who developed and fabricated the Er/Yb glass particles, and Aurelien Clave´, who contributed to the early development of the temperature measurement technique. We thank Deniz Dalkara for providing the GFP viral construct and Valeria Zampini for performing injections. Icons made by Freepik (www.flaticon.com) for graphical abstract and mouse design from Gwilz.
AUTHOR CONTRIBUTIONS
A.P. implemented the Python code into a MATLAB package and combined it with the random phase mask code and performed the simulations. S.D. per-formed in vitro electrophysiological experiments and analyzed the data. C.L. designed the agar phantoms and carried out the temperature measurements. I.-W.C. performed the in vivo electrophysiological experiments and performed virus injections. D.T. designed and built up the scanning-holography system and participated in the in vitro experiments and the temperature measurement.
E.R. and E.P. built up the in vivo holographic system and participated in the
in vivo optogenetics experiments. P.B. and G.T. designed and developed
the temperature measurement technique. D.O. wrote the code for the random phase mask approach. B.C.F. created the theory and developed a first Python code for heat diffusion simulation. V.E. wrote the paper with A.P. and B.C.F. and conceived and supervised the project.
DECLARATION OF INTERESTS
The authors declare no competing interests. Received: August 9, 2017
Revised: March 26, 2018 Accepted: June 28, 2018 Published: July 31, 2018
REFERENCES
Accanto, N., Tanese, D., Ronzitti, E., Molinier, C., Newman, Z.L., Wyart, C., Ehud, I., Eirini, P., and Emiliani, V. (2017). Multiplexed temporally focused light shaping for high-resolution multi-cell targeting. bioRxiv,https://www.biorxiv. org/content/early/2017/11/08/216135.
Adamantidis, A.R., Zhang, F., Aravanis, A.M., Deisseroth, K., and de Lecea, L. (2007). Neural substrates of awakening probed with optogenetic control of hypocretin neurons. Nature 450, 420–424.
Adesnik, H., Bruns, W., Taniguchi, H., Huang, Z.J., and Scanziani, M. (2012). A neural circuit for spatial summation in visual cortex. Nature 490, 226–231.
Aigouy, L., Tessier, G., Mortier, M., and Charlot, B. (2005). Scanning thermal imaging of microelectronic circuits with a fluorescent nanoprobe. Appl. Phys. Lett. 87, 1–3.
Andersen, P., and Moser, E.I. (1995). Brain temperature and hippocampal function. Hippocampus 5, 491–498.
Aravanis, A.M., Wang, L.-P., Zhang, F., Meltzer, L.A., Mogri, M.Z., Schneider, M.B., and Deisseroth, K. (2007). An optical neural interface: in vivo control of rodent motor cortex with integrated fiberoptic and optogenetic technology. J. Neural Eng. 4, S143–S156.
Arias-Gil, G., Ohl, F.W., Takagaki, K., and Lippert, M.T. (2016). Measurement, modeling, and prediction of temperature rise due to optogenetic brain stimu-lation. Neurophotonics 3, 045007.
Aronov, D., and Fee, M.S. (2012). Natural changes in brain temperature under-lie variations in song tempo during a mating behavior. PLoS ONE 7, e47856.
Be`gue, A., Papagiakoumou, E., Leshem, B., Conti, R., Enke, L., Oron, D., and Emiliani, V. (2013). Two-photon excitation in scattering media by spatiotempo-rally shaped beams and their application in optogenetic stimulation. Biomed. Opt. Express 4, 2869–2879.
Beltramo, R., D’Urso, G., dal Maschio, M., Farisello, P., Bovetti, S., Clovis, Y., Lassi, G., Tucci, V., De Pietri Tonelli, D., and Fellin, T. (2013). Layer-specific excitatory circuits differentially control recurrent network dynamics in the neocortex. Nat. Neurosci. 16, 227–234.
Bernstein, J.G., Garrity, P.A., and Boyden, E.S. (2012). Optogenetics and ther-mogenetics: technologies for controlling the activity of targeted cells within intact neural circuits. Curr. Opin. Neurobiol. 22, 61–71.
Bird, R.B., Stewart, W.E., and Lightfoot, E.N. (1976). Transport Phenomena (New York: John Wiley).
Blumm, J., and Lindemann, A. (2003). Characterization of the thermophysical properties of molten polymers and liquids using the flash technique. High Temp. High Press. 35–36, 627–632.
Boulnois, J.L. (1986). Photophysical processes in recent medical laser devel-opments: a review. Lasers Med. Sci. 1, 47–66.
Boyden, E.S., Zhang, F., Bamberg, E., Nagel, G., and Deisseroth, K. (2005). Millisecond-timescale, genetically targeted optical control of neural activity. Nat. Neurosci. 8, 1263–1268.
Brandes, A.R., Elmaklizi, A., Akarc¸ay, H.G., and Kienle, A. (2014). Modeling the tight focusing of beams in absorbing media with Monte Carlo simulations. J. Biomed. Opt. 19, 115003.
Cardin, J.A., Carle´n, M., Meletis, K., Knoblich, U., Zhang, F., Deisseroth, K., Tsai, L.-H., and Moore, C.I. (2009). Driving fast-spiking cells induces gamma rhythm and controls sensory responses. Nature 459, 663–667.
Carslaw, H.S., and Jaeger, J.C. (1947). Conduction of Heat in Solids (Oxford Science Publications).
Chaigneau, E., Ronzitti, E., Gajowa, M.A., Soler-Llavina, G.J., Tanese, D., Brur-eau, A.Y.B., Papagiakoumou, E., Zeng, H., and Emiliani, V. (2016). Two-photon holographic stimulation of ReaChR. Front. Cell. Neurosci. 10, 234. Chen, I.-W., Ronzitti, E., Lee, B.R., Daigle, T.L., Zeng, H., Papagiakoumou, E., and Emiliani, V. (2017). Parallel holographic illumination enables sub-milli-second two-photon optogenetic activation in mouse visual cortex in vivo. bio-Rxiv,https://www.biorxiv.org/content/early/2018/01/21/250795.
Christie, I.N., Wells, J.A., Southern, P., Marina, N., Kasparov, S., Gourine, A.V., and Lythgoe, M.F. (2013). fMRI response to blue light delivery in the naı¨ve brain: implications for combined optogenetic fMRI studies. Neuroimage 66, 634–641.
Crank, J., and Nicolson, P. (1947). A practical method for numerical evaluation of solutions of partial differential equations of the heat-conduction type. Math. Proc. Camb. Philos. Soc. 43, 50–67.
dal Maschio, M., Donovan, J.C., Helmbrecht, T.O., and Baier, H. (2017). Link-ing neurons to network function and behavior by two-photon holographic op-togenetics and volumetric imaging. Neuron 94, 774–789.e5.
Dantelle, G., Mortier, M., Vivien, D., and Patriarche, G. (2005). Nucleation effi-ciency of erbium and ytterbium fluorides in transparent oxyfluoride glass-ce-ramics. J. Mater. Res. 20, 472–481.
De´barre, D., Olivier, N., Supatto, W., and Beaurepaire, E. (2014). Mitigating phototoxicity during multiphoton microscopy of live Drosophila embryos in the 1.0-1.2mm wavelength range. PLoS ONE 9, e104250.
Deng, W., Goldys, E.M., Farnham, M.M.J., and Pilowsky, P.M. (2014). Optoge-netics, the intersection between physics and neuroscience: light stimulation of neurons in physiological conditions. Am. J. Physiol. Regul. Integr. Comp. Physiol. 307, R1292–R1302.
Elwassif, M.M., Kong, Q., Vazquez, M., and Bikson, M. (2006). Bio-heat trans-fer model of deep brain stimulation-induced temperature changes. J. Neural Eng. 3, 306–315.
Emiliani, V., Cohen, A.E., Deisseroth, K., and Ha¨usser, M. (2015). All-optical interrogation of neural circuits. J. Neurosci. 35, 13917–13926.
Ermakova, Y.G., Lanin, A.A., Fedotov, I.V., Roshchin, M., Kelmanson, I.V., Kulik, D., Bogdanova, Y.A., Shokhina, A.G., Bilan, D.S., Staroverov, D.B., et al. (2017). Thermogenetic neurostimulation with single-cell resolution. Nat. Commun. 8, 15362.
Fourier, J. (1822). Fourier: The´orie Analytique de la Chaleur (Chez Firmin Didot, Pe`re et Fils).
Goodman, J.W. (1996). Introduction to Fourier optics McGraw-Hill Series in Electrical and Computer Engineering. Quantum Semiclassical Opt. J. Eur. Opt. Soc. Part B 8, 491.
Guo, Z.V., Hart, A.C., and Ramanathan, S. (2009). Optical interrogation of neu-ral circuits in Caenorhabditis elegans. Nat. Methods 6, 891–896.
Hamada, F.N., Rosenzweig, M., Kang, K., Pulver, S.R., Ghezzi, A., Jegla, T.J., and Garrity, P.A. (2008). An internal thermal sensor controlling temperature preference in Drosophila. Nature 454, 217–220.
Hernandez, O., Papagiakoumou, E., Tanese, D., Fidelin, K., Wyart, C., and Emiliani, V. (2016). Three-dimensional spatiotemporal focusing of holographic patterns. Nat. Commun. 7, 11928.
Hodgkin, A.L., and Katz, B. (1949). The effect of sodium ions on the electrical activity of giant axon of the squid. J. Physiol. 108, 37–77.
Hopt, A., and Neher, E. (2001). Highly nonlinear photodamage in two-photon fluorescence microscopy. Biophys. J. 80, 2029–2036.
Huber, D., Petreanu, L., Ghitani, N., Ranade, S., Hroma´dka, T., Mainen, Z., and Svoboda, K. (2008). Sparse optical microstimulation in barrel cortex drives learned behaviour in freely moving mice. Nature 451, 61–64.
Ji, N., Magee, J.C., and Betzig, E. (2008). High-speed, low-photodamage nonlinear imaging using passive pulse splitters. Nat. Methods 5, 197–202.
Johansson, J.D. (2010). Spectroscopic method for determination of the ab-sorption coefficient in brain tissue. J. Biomed. Opt. 15, 057005.
Kalmbach, A.S., and Waters, J. (2012). Brain surface temperature under a craniotomy. J. Neurophysiol. 108, 3138–3146.
Kedenburg, S., Vieweg, M., Gissibl, T., and Giessen, H. (2012). Linear refrac-tive index and absorption measurements of nonlinear optical liquids in the visible and near-infrared spectral region. Opt. Mater. Express 2, 1588.
Kitamura, T., Ogawa, S.K., Roy, D.S., Okuyama, T., Morrissey, M.D., Smith, L.M., Redondo, R.L., and Tonegawa, S. (2017). Engrams and circuits crucial for systems consolidation of a memory. Science 356, 73–78.
Kiyatkin, E.A. (2007). Brain temperature fluctuations during physiological and pathological conditions. Eur. J. Appl. Physiol. 101, 3–17.
Klapoetke, N.C., Murata, Y., Kim, S.S., Pulver, S.R., Birdsey-Benson, A., Cho, Y.K., Morimoto, T.K., Chuong, A.S., Carpenter, E.J., Tian, Z., et al. (2014). In-dependent optical excitation of distinct neural populations. Nat. Methods 11, 338–346.
Kobat, D., Durst, M.E., Nishimura, N., Wong, A.W., Schaffer, C.B., and Xu, C. (2009). Deep tissue multiphoton microscopy using longer wavelength excita-tion. Opt. Express 17, 13354–13364.
Koester, H.J., Baur, D., Uhl, R., and Hell, S.W. (1999). Ca2+ fluorescence im-aging with pico- and femtosecond two-photon excitation: signal and photo-damage. Biophys. J. 77, 2226–2236.
Ko¨nig, K., Becker, T.W., Fischer, P., Riemann, I., and Halbhuber, K.J. (1999). Pulse-length dependence of cellular response to intense near-infrared laser pulses in multiphoton microscopes. Opt. Lett. 24, 113–115.
Kuhlman, S.J., and Huang, Z.J. (2008). High-resolution labeling and functional manipulation of specific neuron types in mouse brain by Cre-activated viral gene expression. PLoS ONE 3, e2005.
Linz, N., Freidank, S., Liang, X.-X., and Vogel, A. (2016). Wavelength depen-dence of femtosecond laser-induced breakdown in water and implications for laser surgery. Phys. Rev. B 94, 024113.
Lutz, C., Otis, T.S., DeSars, V., Charpak, S., DiGregorio, D.A., and Emiliani, V. (2008). Holographic photolysis of caged neurotransmitters. Nat. Methods 5, 821–827.
Mortier, M., and Patriarche, G. (2000). Structural characterisation of trans-parent oxyfluoride glass-ceramics. J. Mater. Sci. 35, 4849–4856.
Nagel, G., Brauner, M., Liewald, J.F., Adeishvili, N., Bamberg, E., and Gottschalk, A. (2005). Light activation of channelrhodopsin-2 in excitable cells of Caenorhabditis elegans triggers rapid behavioral responses. Curr. Biol. 15, 2279–2284.
Noack, J., and Vogel, A. (1999). Laser-induced plasma formation in water at nanosecond to femtosecond time scales: calculation of thresholds, absorption coefficients, and energy density. IEEE J. Quant. Electron. 35, 1156–1167.
Olivie´, G., Gigue`re, D., Vidal, F., Ozaki, T., Kieffer, J.-C., Nada, O., and Brunette, I. (2008). Wavelength dependence of femtosecond laser ablation threshold of corneal stroma. Opt. Express 16, 4121–4129.
Packer, A.M., Peterka, D.S., Hirtz, J.J., Prakash, R., Deisseroth, K., and Yuste, R. (2012). Two-photon optogenetics of dendritic spines and neural circuits. Nat. Methods 9, 1202–1205.
Packer, A.M., Russell, L.E., Dalgleish, H.W.P., and Ha¨usser, M. (2015). Simul-taneous all-optical manipulation and recording of neural circuit activity with cellular resolution in vivo. Nat. Methods 12, 140–146.
Papagiakoumou, E., de Sars, V., Oron, D., and Emiliani, V. (2008). Patterned two-photon illumination by spatiotemporal shaping of ultrashort pulses. Opt. Express 16, 22039–22047.
Papagiakoumou, E., Anselmi, F., Be`gue, A., de Sars, V., Gl€uckstad, J., Isacoff, E.Y., and Emiliani, V. (2010). Scanless two-photon excitation of channelrho-dopsin-2. Nat. Methods 7, 848–854.
Papagiakoumou, E., Be`gue, A., Leshem, B., Schwartz, O., Stell, B.M., Bradley, J., Oron, D., and Emiliani, V. (2013). Functional patterned multiphoton excita-tion deep inside scattering tissue. Nat. Photonics 7, 274–278.
Pegard, N.M., Oldenburg, I., Sridharan, S., Walller, L., and Adesnik, H. (2017). Three-dimensional scanless holographic optogenetics with temporal focusing (3D-SHOT). Nat. Commun. 8, 1228.
Pennes, H.H. (1948). Analysis of tissue and arterial blood temperatures in the resting human forearm. J. Appl. Physiol. 1, 93–122.
Penzo, M.A., Robert, V., Tucciarone, J., De Bundel, D., Wang, M., Van Aelst, L., Darvas, M., Parada, L.F., Palmiter, R.D., He, M., et al. (2015). The paraventric-ular thalamus controls a central amygdala fear circuit. Nature 519, 455–459.
Petreanu, L., Mao, T., Sternson, S.M., and Svoboda, K. (2009). The subcellular organization of neocortical excitatory connections. Nature 457, 1142–1145.
Pisanello, F., Sileo, L., Oldenburg, I.A., Pisanello, M., Martiradonna, L., Assad, J.A., Sabatini, B.L., and De Vittorio, M. (2014). Multipoint-emitting optical fibers for spatially addressable in vivo optogenetics. Neuron 82, 1245–1254.
Podgorski, K., and Ranganathan, G. (2016). Brain heating induced by near infrared lasers during multi-photon microscopy. bioRxiv 116, 057364.
Pologruto, T.A., Sabatini, B.L., and Svoboda, K. (2003). ScanImage: Flexible software for operating laser scanning microscopes. Biomed. Eng. Online 2, 1–9.
Prakash, R., Yizhar, O., Grewe, B., Ramakrishnan, C., Wang, N., Goshen, I., Packer, A.M., Peterka, D.S., Yuste, R., Schnitzer, M.J., and Deisseroth, K. (2012). Two-photon optogenetic toolbox for fast inhibition, excitation and bi-stable modulation. Nat. Methods 9, 1171–1179.
Reig, R., Mattia, M., Compte, A., Belmonte, C., and Sanchez-Vives, M.V. (2010). Temperature modulation of slow and fast cortical rhythms. J. Neurophysiol. 103, 1253–1261.
Rickgauer, J.P., and Tank, D.W. (2009). Two-photon excitation of channelrho-dopsin-2 at saturation. Proc. Natl. Acad. Sci. U S A 106, 15025–15030.
Rickgauer, J.P., Deisseroth, K., and Tank, D.W. (2014). Simultaneous cellular-resolution optical perturbation and imaging of place cell firing fields. Nat. Neu-rosci. 17, 1816–1824.
Ronzitti, E., Conti, R., Zampini, V., Tanese, D., Foust, A.J., Klapoetke, N., Boy-den, E.S., Papagiakoumou, E., and Emiliani, V. (2017). Sub-millisecond opto-genetic control of neuronal firing with two-photon holographic photoactivation of Chronos. J. Neurosci. 37, 10679–10689.
Saı¨di, E., Samson, B., Aigouy, L., Volz, S., Lo¨w, P., Bergaud, C., and Mortier, M. (2009). Scanning thermal imaging by near-field fluorescence spectroscopy. Nanotechnology 20, 115703.
Schott, S., Bertolotti, J., Le´ger, J.-F., Bourdieu, L., and Gigan, S. (2015). Char-acterization of the angular memory effect of scattered light in biological tis-sues. Opt. Express 23, 13505–13516.
Sharma, H.S., and Hoopes, P.J. (2003). Hyperthermia induced pathophysi-ology of the central nervous system. Int. J. Hyperthermia 19, 325–354.
Shemesh, O.A., Tanese, D., Zampini, V., Linghu, C., Piatkevich, K., Ronzitti, E., Papagiakoumou, E., Boyden, E.S., and Emiliani, V. (2017). Temporally precise single-cell-resolution optogenetics. Nat. Neurosci. 20, 1796–1806.
Shibasaki, K., Suzuki, M., Mizuno, A., and Tominaga, M. (2007). Effects of body temperature on neural activity in the hippocampus: regulation of resting mem-brane potentials by transient receptor potential vanilloid 4. J. Neurosci. 27, 1566–1575.
Shin, Y., Yoo, M., Kim, H.-S., Nam, S.-K., Kim, H.-I., Lee, S.-K., Kim, S., and Kwon, H.-S. (2016). Characterization of fiber-optic light delivery and light-induced temperature changes in a rodent brain for precise optogenetic neuro-modulation. Biomed. Opt. Express 7, 4450–4471.
Stujenske, J.M., Spellman, T., and Gordon, J.A. (2015). Modeling the spatio-temporal dynamics of light and heat propagation for in vivo optogenetics. Cell Rep. 12, 525–534.
Sun, B., Salter, P.S., Roider, C., Jesacher, A., Strauss, J., Heberle, J., Schmidt, M., and Booth, M.J. (2018). Four-dimensional light shaping: manipulating ultra-fast spatiotemporal foci in space and time. Light Sci. Appl. 7, 17117.
Szabo, V., Ventalon, C., De Sars, V., Bradley, J., and Emiliani, V. (2014). Spatially selective holographic photoactivation and functional fluorescence imaging in freely behaving mice with a fiberscope. Neuron 84, 1157–1169.
Thompson, S.M., Masukawa, L.M., and Prince, D.A. (1985). Temperature dependence of intrinsic membrane properties and synaptic potentials in hip-pocampal CA1 neurons in vitro. J. Neurosci. 5, 817–824.
Thomsen, S. (1991). Pathologic analysis of photothermal and photomechan-ical effects of laser-tissue interactions. Photochem. Photobiol. 53, 825–835.
Vogel, A., Noack, J., H€uttman, G., and Paltauf, G. (2005). Mechanisms of femtosecond laser nanosurgery of cells and tissues. Appl. Phys. B Lasers Opt. 81, 1015–1047.
Wang, H., Wang, B., Normoyle, K.P., Jackson, K., Spitler, K., Sharrock, M.F., Miller, C.M., Best, C., Llano, D., and Du, R. (2014). Brain temperature and its fundamental properties: a review for clinical neuroscientists. Front. Neurosci. 8, 307.
Wang, Y., Deng, R., Xie, X., Huang, L., and Liu, X. (2016). Nonlinear spectral and lifetime management in upconversion nanoparticles by controlling energy distribution. Nanoscale 8, 6666–6673.
Wells, J., Kao, C., Konrad, P., Milner, T., Kim, J., Mahadevan-Jansen, A., and Jansen, E.D. (2007). Biophysical mechanisms of transient optical stimulation of peripheral nerve. Biophys. J. 93, 2567–2580.
Wu, Z., Autry, A.E., Bergan, J.F., Watabe-Uchida, M., and Dulac, C.G. (2014). Galanin neurons in the medial preoptic area govern parental behaviour. Nature 509, 325–330.
Wyart, C., Del Bene, F., Warp, E., Scott, E.K., Trauner, D., Baier, H., and Isac-off, E.Y. (2009). Optogenetic dissection of a behavioural module in the verte-brate spinal cord. Nature 461, 407–410.
Yang, W., Carrillo-reid, L., Bando, Y., Peterka, D.S., and Yuste, R. (2018). Simultaneous two-photon optogenetics and imaging of cortical circuits in three dimensions. eLife 7, e32671.
Yaroslavsky, A.N., Schulze, P.C., Yaroslavsky, I.V., Schober, R., Ulrich, F., and Schwarzmaier, H.J. (2002). Optical properties of selected native and coagu-lated human brain tissues in vitro in the visible and near infrared spectral range. Phys. Med. Biol. 47, 2059–2073.
Yizhar, O., Fenno, L.E., Davidson, T.J., Mogri, M., and Deisseroth, K. (2011). Optogenetics in neural systems. Neuron 71, 9–34.
Zhang, Y.-P., and Oertner, T.G. (2007). Optical induction of synaptic plasticity using a light-sensitive channel. Nat. Methods 4, 139–141.
STAR
+METHODS
KEY RESOURCES TABLE
CONTACT FOR REAGENT AND RESOURCE SHARING
Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Valentina Emiliani (valentina.emiliani@parisdescartes.fr).
EXPERIMENTAL MODEL AND SUBJECT DETAILS Mice forin vivo experiments
All animal experiments were performed in accordance with the Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010. The protocols were approved by the Paris Descartes Ethics Committee for Animal Research with the registered number CEEA34.EV.118.12. Adult female or male C57BL/6J mice (Janvier Labs) were anesthetized with intraperitoneal injection of a ketamine-xylazine mixture (0.1 mg ketamine and 0.01 mg xylazine/g body weight) during stereotaxic injection and with isoflurane (2% for induction and 0.5%–1% for experiment) during photostimulation experiments. Cortical neurons of 4-week-old mice were transduced with viral vectors of opsins using stereotaxic injection. Photostimulation experiments were performed 5-8 weeks after injection.
Mice for in-vitro experiments
All experimental procedures were conducted in accordance with guidelines from the European Union and institutional guidelines on the care and use of laboratory animals (council directive 86/609 of the European Economic Community) that were approved by the Paris Descartes Ethics Committee for Animal Research (registration number CEEA34.EV.118.12). Stereotactic injections of the viral vectors AAV2/8-hSyn-CoChR-GFP (Shemesh et al., 2017) were performed in 4-week-old male Swiss mice (Janvier Labs). Animals were housed from 3 to 5 per cage with a light dark cycle of 12 + 12 h. Mice were anesthetized with a ketamine (80 mg/kg)-xylazine (5 mg/kg) solution and a small craniotomy (0.7 mm) was made on the skull overlying V1 cortex. An injection of 1–1.5ml solution con-taining the viral vector was made with a cannula at about 80–100 nl/min at 200–250mm below the dural surface. The skin was sutured, and the mouse recovered from anesthesia.
METHOD DETAILS
Virus injection and surgical procedures – In-vivo
Through a craniotomy over the right primary visual cortex (V1; 3.5 mm caudal from the bregma, 2.5 mm lateral from the midline), 1.5mL viral vectors AAV2/8-hSyn-CoChR-GFP (Klapoetke et al., 2014; Shemesh et al., 2017) of were delivered via a cannula in
REAGENT or RESOURCE SOURCE IDENTIFIER Bacterial and Virus Strains
AAV2/8-hSyn-CoChR-GFP Klapoetke et al., 2014 N/A Experimental Models: Cell Lines
Mouse for in-vivo: C57BL/6J Janvier Labs SC-C57J-M Mouse for in-vitro: Swiss mice Janvier Labs RjOrl:SWISS Software and Algorithms
pCLAMP10 Molecular Devices http://mdc.custhelp.com/
ScanImage 3 Pologruto et al., 2003 http://scanimage.org
Holographic and Scanning illumination thermal simulation (MATLAB)
This paper Available as a Supplemental file MATLAB R2017a Mathworks inc., USA http://mathworks.com
LabVIEW National Instruments http://www.ni.com/en-us.html
OriginPro 2016 OriginLab https://www.originlab.com/Origin
AvaSoft8 Avantes https://www.avantes.com
SlideBook 6 3i-Intelligent Imaging Innovations https://www.intelligent-imaging.com/slidebook
Wavefront Designer 4 Photonics Department, Institut de la Vision, Paris
the layer 2/3 (250mm deep) at a speed of 80-100 nL/min. For performing acute photostimulation in vivo, a circular craniotomy of 2 mm diameter was made over V1 and the dura mater was removed. Agarose of 0.5%–2% and a cover glass were applied on top of the craniotomy for stabilization/to dampen tissue movement.
2P-guided electrophysiology – in-vivo
Cortical neurons were targeted with patch pipettes under a custom-built 2P microscope equipped with a Ti:Sapphire laser (Chameleon Vision II, Coherent), and a 40x water-immersion objective (Nikon, CFI APO 40XW NIR, NA0.80). For a detailed descrip-tion of the 2P scanning imaging system see (Chaigneau et al., 2016). The GFP labeling in opsin-expressing cells were visualized by excitation at 920 nm and the emitted fluorescence was collected through red (617/70 nm) and green (510/80 nm) filters (Semrock). Imaging data were acquired using ScanImage 3 software (http://scanimage.org).
Cell-attached recordings were obtained by using microelectrodes fabricated from borosilicate glass (5-8 MU resistance) and filled with solution containing the following (in mM): 135 potassium gluconate, 10 HEPES, 10 sodium phosphocreatine, 4 KCl, 4 Mg-ATP, 0.3 Na3GTP, 25-50 Alexa Flour 594 for pipette visualization. The craniotomy was covered with the extracellular solution containing the
following (in mM): 145 NaCl, 5.4 KCl, 10 HEPES, 1 MgCl2, 1.8 CaCl2. Voltage recordings in the current-clamp mode were acquired
by using a MultiClamp 700B amplifier and a Digidata 1550A digitizer, which were controlled by a pCLAMP10 software (Molecular Devices). Electrophysiology data were filtered at 6 kHz and digitized at 20 kHz.
Holographic photostimulation – in-vivo
Holographic photostimulation was implemented with the imaging system mentioned above. Computer-generated holography was utilized for patterning light beams from an amplified fiber laser (Satsuma HP, Amplitude Systemes) at 1030 nm via a spatial light modulator (LCOS-SLM; X10468-07, Hamamatsu Photonics). The photostimulation setup was similar to the one described in (Ronzitti et al., 2017) and (Chaigneau et al., 2016). The SLM was controlled by a custom-designed software (Lutz et al., 2008). A cylindrical lens was introduced to suppress zero-order excitation (Hernandez et al., 2016).
2P photostimulation was performed using a circular holographic spot of 12-mm diameter covering the soma of a CoChR-positive neuron whose spiking activity was monitored through a patch pipette. A threshold power density between 0.05-0.5 mW/mm2of a 1-10 ms light pulse was determined for a target neuron to elicit an AP in 3-6 repetitions. A train of APs were elicited upon photosti-mulation with 5-10 illumination pulses at 10, 20 and 40 Hz.
Brain slices
Brain slices of V1 cortex were prepared from mice 7–15 weeks after viral injection. Mice were deeply anesthetized with isoflurane (5% in air) and decapitated, and the brain was rapidly removed. Sagittal slices 300mm thick were obtained (VT1200S Leica Biosystems, Germany) in room temperature or ice-cold solution containing the following (in mM): 93 NMDG, 2.5 KCl, 1.25 NaH2PO4, 30 NAHCO3, 20 HEPES Acid, 25 Glucose, 2 thiourea, 5 Na-Ascorbate, 3 Na-Pyruvate, 0.5 CaCl2 and 10 MgCl2. Afterward, slices were transferred to a recovery chamber held at 35for 45 min, in a bath containing the following (in mM): 125 NaCl, 2.5 KCl, 26 NaHCO3, 1.25 NaH2PO4,
1 MgCl2, 1.5 CaCl2, 25 glucose, 0.5 ascorbic acid. All solutions were aerated with 95% O2and 5% CO2to a final pH of 7.4. Slices were
placed in a recording chamber under the microscope objective and were patched while monitoring IR transmitted light images acquired at approximately video rate. Cells were patched at 40–70mm depth and clamped at 70 mV in voltage-clamp configuration. Opsin expressing cells were identified via epifluorescence or 2P-scanning imaging.
Whole-cell recordings – in-vitro
Whole-cell patch clamp recordings were made using Multiclamp 700B amplifier and a Digidata 1440 digitizer and a PC running pClamp (Molecular Devices). Cell type was established based on morphology and AP firing properties. Membrane potential was kept at70 mV with current injections ranging from 5 to 35 pA in current-clamp configuration. Voltage and current clamp record-ings were filtered at 6–10 kHz and sampled at 20–50 kHz. Borosilicate glass pipettes (outer diameter 1.5 mm and inner diameter 0.86 mm) were pulled with a micropipette puller (Sutter Instruments) and filled with a solution containing the following (in mM): 130 potassium gluconate, 7 KCl, 4 Mg-ATP, 0.3 mM Na-GTP, 10 sodium phosphocreatine and 10 mM HEPES (pH adjusted to 7.28 with KOH; osmolarity 280 mOsm). Pipette resistance in the bath was 5–7 MU.
2P holographic and spiral scanning photostimulation – in-vitro
The optical system used is analogous to the one described above for the measure of the laser induced heating. In this experiment, the excitation laser (GOJI, Amplitude Systemes, here operated at a repetition rate of 2 MHz and laser pulse width of300fs) could be alternatively directed onto two different optical paths, in order to generate either parallel holographic stimulation or spiral scanning stimulation. The former consisted in illuminating the whole cell body with a 15mm diameter holographic spot (see description of the optical path in the Thermal measurement section). The latter consisted in the spiral scanning of the cell body with an almost diffraction limited spot (0.8 mm lateral size). The spiral trajectory was obtained by controlling the movement of the two galvomirrors using a SlideBook 6 commercial software (3i-Intelligent Imaging Innovations) and its parameters (7 tours, pitch 1mm) were chosen to cover approximately the same surface of the holographic illumination. The two beams were focused on the sample by a 40x 0.8 NA water immersion Zeiss objective and the cell response to photostimulation was monitored by electrophysiological recording.
![[PDF] Apprendre à compiler un programme MATLAB | Formation informatique](data:image/gif;base64,R0lGODlhAQABAIAAAP///wAAACH5BAEAAAAALAAAAAABAAEAAAICRAEAOw==)